Hydrobiologia 448: 71–81, 2001. © 2001 Kluwer Academic Publishers. Printed in the Netherlands.
71
Microbial community structure and function in response to larval chironomid feeding pressure in a microcosm experiment P. E. Yeager1 , C. L. Foreman2 & R. L. Sinsabaugh3 1 School of Science and Math, Fairmont State College, Fairmont, WV 26554, U.S.A. Tel.: +1-304-363-2683. Fax: +1-304-363-4304. E-mail:
[email protected] 2 Department of Land Resources and Environmental Sciences, Montana State University, Bozeman, MT 59717, U.S.A. 3 Environmental Science Department, University of Toledo, Toledo, OH 43606, U.S.A.
Received 24 November 1998; in revised form 9 January 2001; accepted 7 February 2001
Key words: chironomid, feeding, microbial, enzyme, nutrient, molecular, microcosm
Abstract Studies concerning the interactions between macroinvertebrate and microbial communities have been carried out for some time. However, most of these studies have been simple feeding relationships that as a group have produced ambiguous results. We perceive these relationships to be more complex, encompassing not only microbial population density but also structure and function. To further understand these relationships, we employed molecular and biochemical techniques to study microbial structural and functional diversity in relation to macrobenthic feeding pressure. We studied the effect of feeding by the three population densities of larval midge, Chironomus tentans, (Diptera: Chironomidae) on microbial community organization. No significant difference in microbial biomass carbon (10.0 mg/g DWS ±1.97) was seen between the three treatments. However, we did detect significant shifts in microbial structure and function with increases in midge population density. The activities of carbon (C) and nitrogen (N) acquiring enzymes were negatively correlated with midge population density. While the C:N ratio was positively correlated with midge population density, suggesting that while the concentration of nitrogen decreased, its availability to the chironomids increased. There was also a marked difference in microbial community structure with increasing midge population density. These shifts in microbial organization are indicative of a complex set of interactions between the microbial community and the chironomid larvae.
Introduction The relationship of macroinvertebrates and microbial communities was first reflected on by Davies (1925) when he noted “It is not improbable that the association of various groups of animals with soils of particular texture is not so much influenced by the texture per se as by the food conditions of which the size of the soil particles is a correlated indication”. ZoBell & Feltham (1938) demonstrated that bacteria could be used as a nutritional source by invertebrates, and that bacterial densities in marine sediments indicated that their contribution could be significant. Beyond this, they speculated that by metabolizing dissolved organic matter (DOM), the microbial community made this
otherwise inaccessible resource available to invertebrates. The importance of the microbial community to invertebrate nutrition was further clarified when Rodina (1949) observed that detritus could not be readily digested by chironomids without conditioning by bacteria. He also speculated that chironomids feeding on algae were not deriving nutrition from the algae, but from the associated bacteria. In a laboratory experiment, Rodina (1949) showed chironomids could complete their life cycle with bacteria as their sole larval food source. Hilsenhoff (1966), while studying the population ecology of Chironomus plumosus (Diptera: Chironominae) in Lake Winnebago, Wisconsin, observed that the regulation of this community
72 was a complex interaction of many factors and speculated that: “unknown viruses, fungi and bacteria were probably the most important regulators of the population”. The importance of the microbial community as a nutritional source was contested, when Baker & Bradnam (1976) calculated that only 7% of the microbial community was assimilated by Chironomidae. However, they theorized that the limited digestion of the microbes yielded essential nutrients that were otherwise unavailable to the chironomid community. Several authors supported the idea that low microbial biomass and restricted digestibility of the microbes limited their significance in chironomid nutrition (Cammen, 1980; Findlay et al, 1984; Kemp, 1987; Johnson et al., 1990). In contrast, however, many investigators reached the opposite conclusion. Ward & Cummins (1979) defined invertebrate food quality as an increase in microbial biomass per unit of detritus. Fry (1982) concluded that as much as 82% of the microbial community was digested by chironomids, making them a significant source of carbon. In comparing the importance of organic matter (OM) quality to other sediment parameters effecting chironomid distribution, Johnson (1984) concluded that OM quality of which microbial biomass was a correlated indication was the most important parameter. Similar descriptions of invertebrate microbial interaction were discussed by several authors (Mclachlan & Dickson, 1977; Toscano & McLachlan, 1980; McGarrigle, 1980; McMurtry, 1983; Carmen & Thistle, 1985; Reynoldson, 1987). The sum of this research is that no direct relationship has been established between the microbial and invertebrate communities. Two other concepts pertaining to invertebrate nutrition were also brought forward during the 1980s and 1990s. The first is that of direct adsorption of DOM by invertebrates. This route of carbon accumulation was proposed by Hershey (1996) for Simulìidae (Insecta: Diptera) and while promising requires further research. The second concept is that mucilage or biofilms produced by DOM flocculation, or microbial in nature, may be an important food source for invertebrates (Hobbie, 1980; Johnson et al., 1989; Hax & Golliday, 1993). This concept in freshwater ecosystems has not been pursued. However, in the marine environment, this line of research has demonstrated the use of these substrates by several invertebrates (Taghon, 1982; Baird & Thistle, 1986; Bärlocher &
Murdoch, 1989; Phillips & Kilambi, 1994; Decho, 1996). None of the above research is concerned with the ecology of the microbial community in response to invertebrate feeding. Hargrave (1970) used the amphipod Hyalella azteca to look at the effect of feeding pressure on benthic microflora. He was able to demonstrate increased bacterial respiration at population densities of H. azteca of up to four times those naturally occurring in the sediments. Above this feeding pressure, bacterial respiration was seen to decline. Hargrave suggested that increases in bacterial respiration could be due to increased surface area, fecal input from H. azteca, or the cropping of senescent or antibiotic producing cells. In an experiment using macrobenthic organism representing three different functional feeding groups, Van De Bund et al. (1994) found that they could predict the affect on microbial communities by knowing about the feeding ecology of the macrobenthic organism concerned. The organisms studied were: Monoporeia affinis, an amphipod and nonselective sediment detritus feeder; the tube building midge larvae Chironomus riparius a surface deposit feeder; and the Oligochaete Tubifex tubifex a deposit feeder. They found that only the Oligochaete negatively influenced bacterial abundance, and assumed this was because of feeding specificity. However, when microbial productivity was measured by incorporation of tritiated thymidine, there was an increase in bacterial productivity under each of the feeding regimes. The greatest increase in productivity was found in the chironomid experiment, and the authors concluded that the increase in bacterial production for all three macrobenthic feeding groups was due to bioturbation of the sediments. Traunspurger et al. (1997) studying the relationship between nematodes and the bacterial community concluded that direct bacterial foraging by the nematode Caenorhabditis elegans and not bioturbation was the cause for increased microbial activity as measured by tritiated thymidine. Both of these studies conclude that the macrobenthos can be correlated to increases in microbial activity. However, neither these studies nor those concerning microbes as nutrient sources for the macrobenthos are concerned with the ecology of the microbial community. The research conducted here was designed to study the ecological relationship between grazing by Chironomus tentans (Diptera: Chironomidae) larvae and microbial community organization and may be have relevance to other collector/gatherer macroinverteb-
73 rates. Microbial community structure and function were assessed through biochemical and molecular techniques measuring microbial community heterogeneity, biomass, substrate utilization and enzyme activity in reference larval population density.
Materials and methods Microcosms Nine microcosms were constructed from 300 ml beakers with a surface area at the bottom of 41.85 cm2 . Each beaker contained a 0.1 g dry weight (d/w) brown paper towel disk covering the bottom of the beaker and 0.25 g d/w shredded toweling substratum. Shredding was accomplished using a blender to produce greater surface area, but not high fiber content. When wetted toweling filled the beakers to an approximate height of 2 cm. Paper towels used in the experiment were boiled and rinsed three times before use in the microcosms. In order to inoculate the microcosms with a similar microbial community, each beaker was initially filled with 300 ml of water from the culture tank, aerated and allowed to equilibrate for 3 days. Thereafter, dechlorinated tap water was added to maintain the volume of water at 300 ml for the duration of the experiment. Each of the microcosms were oxygenated during the entire experiment and DO along with temperature were monitored. Second instar C. tentans larvae, determined by body length, head capsule size and color (Townsend et al., 1981), were added to six of the nine microcosms. Three of the microcosms received no larvae, three had five larvae added to each beaker and three beakers each contained ten larvae. This produced an array of nine beakers three with no larvae, three with five per beaker and three with ten larvae per beaker. Tetra Min condition food was suspended in deionized water at a concentration of 0.02 g ml−1 and 0.5 ml was delivered to each microcosm daily. After 14 days, the larvae were removed and the remaining substratum frozen in liquid nitrogen then transferred to a −80 ◦ C freezer. DNA extraction and hybridization DNA extraction was based on the method of More et al. (1994). This DNA was used to conduct whole community DNA–DNA hybridizations. Through DNA– DNA hybridizations, microbial communities can be compared without the need for culturing colonies in the lab (Lee & Fuhrman, 1991). This is accomplished
by extracting the DNA from the organism in different communities, and denaturing it into single stranded DNA (sDNA). The sDNA from different communities is then brought back together as described below and allowed to anneal with sDNA from different microbial communities. The degree to which the sDNAs anneal is then used to measure similarity between the different communities. To extract microbial DNA, separate substratum samples from each of the treatment microcosms of 0.5 g were added to 2 ml microcentrifuge tubes containing 0.5 g of 0.1 mm zirconia/silica beads. To each tube, 500 μl of phosphate buffer (100 mm, pH 8, and 250 μl of lysis solution (10% SDS/500 mm Tris/100 mM NaCl) were added, and the tubes capped. The contents of the tubes were then homogenized for 2 min, 30 s on then 30 s off, using a Mini bead-beater/8 (Biospec; Tulsa, OK), kept in a 5 ◦ C cold room to minimize frictional heating. After milling, the tubes were transferred to an ice bath. The tubes were then spun at 10 000 × g for 1 min and the supernatant removed. An additional 500 μl buffer solution was then added to the tubes and samples were placed in the bead beater for 30 s to bring the remaining nucleic material into solution. Tubes were then spun at 10 000 × g and the resulting supernatant added to that already collected for each tube. Each aliquot was then divided into two 350 μl subsamples and the non-DNA component precipitated with 7.5 M ammonium acetate (5:2). DNA was removed from the supernatants using a Wizard miniprep system (Promega; Madison, WI). DNA fragment length was verified by agarose gel electrophoreses using a Lambda DNA/EcoRI, Hind 111 ladder (Promega; Madison, WI), and ethidium bromide staining. DNA was quantified with PicoGreen dsDNA quantitation reagent (Molecular Probes, Eugene, OR) at excitation and emission wavelengths of 480 nm and 520 nm, respectively. Whole community DNA probes were produced through random primed labeling using the 32 P and the multiple prime labeling system (Amersham, U.K.). One probe was constructed from each of the 3 chironomid densities (0, 5 and 10 per microcosm). These whole community probes allow for the comparison of microbial community structure by ‘finger printing’ each community through the DNA present. The microbial community structure between each treatment can then be compared by affixing s DNA from one treatment to a membrane and allowing the single stranded probe labeled with 32 P to anneal to it. Similarity
74 between treatment microbial communities can then be assessed through hybridization and relative annealing measured through autoradiography of the membranes. Hybridizations were conducted by dot blotting triplicate 20 μg sDNA samples from each of the nine microcosms along with a negative control series of herring sperm DNA, on to a positively charged nylon membrane (Bio-Rad, Hercules, CA). This produced a 9 × 9 target matrix with one row of serially diluted herring sperm as a negative control. The sDNA was fixed to the membrane by exposing each side of the membrane to 120 000 μJ of UV energy using a Stratlinker (Stratagen; La Jolla, CA). The membranes were then crosslinked thermally by baking at 80 ◦ C for 1 h. Optimum prehybridization, hybridization and wash conditions were established in preliminary trials. Hybridizations were run in standard buffer (Bio-Rad, Hercules, CA), at 65 ◦ C and reactions ran for 14–16 h. After hybridization, the membrane was washed then placed on BioMax-MS radiography film (Kodak, NY). The developed films were scanned using a flat bed scanner and dot intensities quantified using Quantity one image analysis software (PDI, NY). Microbial biomass estimation Microbial biomass was measured as lipid-P (μg g-1 ash-free dry mass (AFDW)) following the method of White et al. (1979). Microbial phospho-lipid (lipid-P) was extracted from approximately 1 g (w/w) substratum with an extraction solution of 50 mM phosphate buffer, methanol and chloroform in a 7:30:15 (v/v/v) ratio. After the edition of the extraction solution, samples were sealed and placed in a 30 ◦ C dry bath for 2 h. Tubes were then spun at 500 × g to separate phases, and 2.0 ml of the chloroform layer transferred to a 15 ml acid washed glass test tube. The glass test tubes were transferred to an evaporation manifold and the chloroform evaporated. To increase evaporation efficiencies, nitrogen was passed through tubes at a flow rate of 100 ml min−1 and the manifold brought to 60 ◦ C. To hydrolyze the phospholipid 1.0 ml of 10 M H2 SO4 and 0.25 ml of 30% H2 O2 were added to each tube and the dri-bath temperature raised to 180 ◦ C for 2.0 h, or until the solution turned clear. This solution was neutralized with 5 M NaOH and the phosphate content quantified by the ascorbic acid method (Standard Methods, 1992). For conversion from lipidP to microbial biomass carbon (MBC), a coefficient
of 191.7 μg of carbon per 100 ηM lipid-P was used (Findlay et al., 1989). Substrate utilization Substrate utilization, using Biolog microplates, is a method by which to conduct community level physiological profiling of special and temporal variation in gram-negative bacterial communities (Garland & Mills, 1991; Garland, 1997). Gram-negative microplates (GN MicroPlateþ) (Biolog, Inc. Haywood, CA) were used to determine substrate utilization by the microbial community (Zak et al., 1994; Garland & Mills, 1997). GN MicroPlates (GN-plate), 96 well plates containing 95 different carbon substrates and a negative control. Each of the 96 wells contains tetrozolium violet, a redox dye, which is reduced to formazan by microbial respiration. The amount of substrate utilization is measured by the intensity of the color change within each well. Relative respiration rates are measured by monitoring optical density at 590 ηm. For analysis purposes, the 95 separate substrates from the GN-plates have been collapsed into 6 categories (Amines-amides, amino acids, polymers, carbohydrates, carboxylic acids and miscellaneous). To analyze substrate utilization, we used 0.5 g of paper towel from each of the nine microcosms. The paper toweling was homogenized for 10 s in 50 mM acetate buffer using a polytron. Samples were then stored on ice until ready for use. Prior to placement in the GN-plate samples were resuspended in the buffer by shaking and then placed on a stir plate to maintain suspension. Samples were loaded using an eight-channel pipetter and triplicate plates were run for each sample. Samples were incubated at 25 ◦ C for 24 h and optical densities read at 590 ηm. Enzyme activity The microbial community deploys extracellular enzymes to degrade specific substrates and transport them into the cell. Production of the enzymes is presumed to be metabolically expensive, but is controlled by feedback mechanisms to increase efficiency. The activity of these enzymes can be monitored in the lab through the use of specific substrates thereby indicating relative availability of each substrate in the environment (Sinsabaugh et al., 1999). We used enzyme substrates linked to methylumbelliferyl (MUB) and methyl coumarin (MCA) to determine enzyme activities for 22 different substrates.
75 The fluorescent moieties, MCA and MUF, are similar (biphenyl) compounds which are covalently bonded to parent substrate compounds and only become fluorescent when the bonds are broken. The bond between MCA and L-leucine is peptide like, whereas the bond between MUF and β-D-gluc is similar to the β-D-glycosidic linkages of cellobiose and cellulose. Therefore, these substrates will be hydrolyzed by aminopeptidase and β-D-glucosidase enzymes respectively (Zak, 1994). Each of the substrates and controls were replicated three times on each plate. Each of the wells received 50 μl of substrate, and 200 μl of sample and were allowed to incubate at room temperature for 2 h. Fluorescence was then read at an excitation wavelength of 365 nm and emission wavelength of 450 nm and converted to enzyme activity in nmol h −1 ml−1 . Organic matter content Organic matter (OM) content of the substratum was determined by the loss on ignition (LOI) method (Dean, 1974). Paper towel samples from each of the nine microcosms were weighed and then dried at 50 ◦ C for 48 h. Dry weight was then determined and the samples ashed in a muffle furnace at 500 ◦ C for 2 h. The OM content of each sample was then calculated as the difference between d/w and the ash weight (AFDW). Statistical analysis Tukey’s HSD and cluster analysis were performed using Statistica 4.0 (StatSoft; Tulsa, OK). For the Tukey’s analysis, the alpha level was set at 0.05. For cluster analysis, data from all nine microcosms was z-transformed so that the magnitude of any data point would not unduly weight the dissimilarity matrices. Cluster analysis employed squared euclidean distances and Ward’s agglomeration method (Ward, 1963) to obtain measures of dissimilarity between the three treatments.
Results Observations of microcosms and chironomids All chironomids in the 5 and 10 chironomid microcosms were observed to have matured during the duration of the experiment and no other organisms
(nematodes, fungi, etc.) were observed in the microcosms. Most of the paper toweling that had been introduced to the microcosms showed evidence of having been worked by the chironomids and developed into tubes, or was ingested, and represented as frass. There was relatively more unworked paper toweling remaining in the 5-chironomid microcosms than the 10. Paper toweling in the zero chironomid microcosms showed no evidence of macroinvertebrate working, but did appear to be somewhat degraded over the duration of the experiment. This was presumably due to microbial degradation of the cellulose. Dissolved oxygen and temperature was monitored daily in each of the microcosms and ranged from 7.93 to 8.61 ppm and fro 23 to 25 ◦ C. Microbial biomass The mean microbial biomass carbon (MBC) in the microcosms with 0 and 5 chironomids was 8.9 mg g−1 AFDW ±1.24 and 8.8 mg g−1 AFDW ±2.3, respectively. In the microcosm with 10 chironomids, the MBC was 12.3 mg g−1 AFDW ±4.9. There was no statistically significant difference (α = 0.05) between the means of MBC in three treatments (0, 5 and 10 chironomids). Enzyme activities Of the 22 enzyme substrates studied, 17 were acted upon by the microbial community (Table 1). These enzymes could be further differentiated by the pattern of activity across the 3 microbial densities. They produced 4 patterns relating to chironomid population density: increasing activity, decreasing activity, highest activity at middle population density and lowest activity at middle population density (Table 1). Eleven of these extracellular enzymes had significantly different (α = 0.05) activities as indicated by Tukey’s analysis between at least 2 of the 3 chironomid densities (Table 2). Five of the enzyme activities (β-glucuronidase, β-glucosidase, βgalactosidase, β-glucosidase, leucine-aminopeptidase and serine) were significantly different between each of the 3 chironomid densities. For glutamic acid and glycine, there was a significant difference in activity between the microcosms with 0 and 10 chironomids. Enzyme activities associated with arabinopyranoside and cellobiohydrolase were not significantly different between the microcosms with 0 and 5 chironomids, but each were significantly different than the microcosm with 10 organisms. For
76 Table 1. Mean and standard deviation (SD) of microbial enzyme activities (μmol h−1 g−1 AFDW) for C. tentans population densities of 0, 5 and 10 organisms per microcosm Population density Substrate Proline Glutamic acid Tyrosine Sulfatase β-Xylosidase Alanine Glycine Serine Leucine aminopeptidase Cellobiohydrolase α-Glucosidase Arabinopyranoside β-Glucuronidase β-Galactosidase β-Glucosidase N-Acetly-glucosimine Phosphatase Guanidinobenzoate Pyroglutamic acid Araginie Aspartic acid Asparagine
0
5
Mean
SD
Mean
SD
0 231 0 1510 15 310 726 3032 3364 5777 7053 1506 1144 3933 6653 10 922 11 845 0.0 0.0 0.0 0.0 0.0
0 206 0 279 26 335 182 34 361 698 142 474 55 420 596 9466 950 0.0 0.0 0.0 0.0 0.0
160 400 252 4183 0 0 596 1508 2445 5270 5434 1566 4488 5443 12 697 35 4632 0.0 0.0 0.0 0.0 0.0
223 42 295 602 0 0 43 225 162 1018 296 81 580 110 2842 61 8022 0.0 0.0 0.0 0.0 0.0
sulfatase, there was a significant difference between the microcosm with 0 organisms and the two higher densities. β-glucosidase activity was significantly different between the 0 and 5 chironomid densities, but not between the 0 and 10 or 5 and 10 densities. When categorized as either nitrogen acquiring or carbon acquiring enzymes, and taken as a grand mean of those activities, these two groupings of enzyme activity displayed distinct trends in microbial function with increasing chironomid population density. These means were not statistically significantly different (α = 0.05) between treatments, but may have been ecologically significant. Overall activity of both enzyme systems had a high negative correlation with increasing chironomid population density and the ratio of enzyme activity between the carbon and nitrogen demonstrated a positive correlation with chironomid population density (Table 3).
10 Mean 439 591 656 4333 0 0 91 375 1160 1726 2508 753 2840 2225 8490 13 331 6507 0.0 0.0 0.0 0.0 0.0
SD 760 79 577 881 0 0 79 202 78 305 168 50 894 104 1313 1002 6530 0.0 0.0 0.0 0.0 0.0
Substrate utilization Each of the 6 GN-plate substrate categories (amines/ amides, amino acids, carbohydrates, carboxylic acids, miscellaneous and polymers) supported microbial respiration during the 24 h incubation. Of the three chironomid population densities studied (0, 5 and 10) microbial respiration was highest in the microcosms with 5 chironomids for each of the substrate categories (Table 4). Each of the 6 categories demonstrated significantly different (α = 0.05) respiration between at least 2 of the 3 chironomid treatment densities. For all of the categories, respiratory activity in the microcosm with 5 individuals was significantly higher (α = 0.05) than in the microcosm with 10 chironomids, but not than the microcosm with 0 chironomids (Table 5). Community structure Whole community DNA–DNA hybridization analysis of microbial community structure indicated a 79% dis-
77 Table 2. Analysis of variance for microbial enzyme activities with a significant difference (α = 0.05) between treatments using Tukey’s multiple range test at C. tentans population densities of 0, 5 and 10 organisms per microcosm. Overlapping bars indicate no significant difference in enzyme activity between treatments Substrate
Glutamic Acid
Sulfatase
Glycine
Serine
Leucine Aminopeptidase
Cellobiohydrolase
Density 0 5 10 0 5 10 0 5 10 0 5 10 0 5 10 0 5 10
α-Glucosidase
0 5 10
Arabinopyranoside
0 5 10
β-Glucuronidase
0 5 10
β-Galactosidase
0 5 10
β-glucosidase
0 5 10
Table 3. Grand means and correlation coefficients for carbon and nitrogen acquiring microbial enzyme activities (μmol h−1 g−1 AFDW) and the ratio between the two at C. tentans population densities of 0, 5 and 10 organisms per microcosm Density
Carbon (r = −0.99)
Nitrogen (r = −0.99)
Ratio (r = 0.98)
0 5 10
4137 3926 3607
1239 827 454
3.3 4.7 8.0
Tukey’s ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗
Table 4. Mean and standard deviation (SD) of substrate utilization as optical density for the microbial community at C. tentans population densities of 0, 5 and 10 organisms per microcosm
∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗
Population density Substrate
Mean
0 SD
5
Amines/Amides Amino acids Carbohydrates Carboxylic acids Miscellaneous Polymers
12.2 57.5 77.1 38.8 26.8 23.8
1.9 11.1 14.2 9.2 2.3 7.3
Mean 16.7 70.4 112.5 49.5 35.9 30.6
SD
10 Mean SD
4.4 7.1 17.5 30.0 8.3 48.5 19.3 13.1 8.8 13.5 0.8 12.3
2.2 8.3 22.6 6.8 7.7 5.6
∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗
similarity using cluster analysis between the 0 and 5 chironomid treatments. The microbial community structure in the microcosm with 10 chironomids was 100% dissimilar to that of the microcosms with 5 and 0 chironomids as determined by cluster analysis.
∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗
Discussion Through the use of contemporary microbial ecology methods, this study was able to demonstrate specific shifts in a microbial structure and function as a result of feeding pressure by Chironomus tentans. These included: increased availability of C and N to the microbial community, trends in substrate utilization by gram-negative bacteria related to chironomid population density, and changes in microbial community structure with increased feeding pressure by chironomids. This study did not demonstrate an effect on microbial biomass with increased chironomid feeding pressure. Ecologically, this experiment provides greater insight into microbial structure and function than observed by earlier researchers (Hargrave, 1970;
78 Table 5. Analysis of variance using Tukey’s multiple range test of microbial substrate utilization (GN plate) in optical density for C. tentans population densities of 0, 5 and 10 organisms per microcosm. Overlapping bars indicate no significant difference in substrate utilization between treatments (α = 0.05) Substrate
Density
Tukey’s
Amines/Amides
0 5 10
∗∗∗∗∗∗
0 5 10
∗∗∗∗∗∗
Amino acids
Carbohydrates
Carboxylic acids
Miscellaneous
Polymers
∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗
proposed as an enriched substrate for microbial populations (Hargrave, 1976), but although the amount of frass observed increased with increasing chironomid population density, no corresponding increase in MBC was found. These results are not unlike those found in the literature, where when taken as a whole, point to no direct relationship between the microbial community biomass and invertebrate activity (Mclachlan & Dickson, 1977; Toscano & McLachlan, 1980; Carmen & McGarrigle, 1980; Cammen, 1980; McMurtry, 1983; Findlay et al., 1984; Thistle, 1985; Reynoldson, 1987; Kemp, 1987; Johnson et al., 1990).
∗∗∗∗∗∗
Enzyme activity ∗∗∗∗∗∗
0 5 10
∗∗∗∗∗∗
0 5 10
∗∗∗∗∗∗
0 5 10
∗∗∗∗∗∗
0 5 10
∗∗∗∗∗∗
∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗ ∗∗∗∗∗∗
Van De Bund et al., 1994; Traunspurger et al., 1997) among others. Microbial biomass When considering the relationship between the chironomid and microbial communities using the conventional metric of microbial biomass, there was no statistically significant difference (α = 0.05) in microbial biomass between three treatments. Nominally, microbial biomass was greatest in the microcosms with 10 chironomids, however this difference was the result of a single outlier in one of the replicates. While a stimulatory affect on microbial production has been proposed by some authors, as a result of substrate bioturbation and feeding (Johnson et al., 1989; Goedkoop, 1992), there is no evidence to support that theory in this study. The use of frass has also been
Several patterns of microbial enzyme activity were evident with increases in chironomid population density. Two enzyme activities in particular show a direct relationship with chironomid population density Proline and Tyrosine. There is no activity for these enzymes in the 0 chironomid treatment, and an increase in activity for each enzyme with increasing chironomid population density. This demonstrates ecological significance to the microbial community although there was no statistical significance (α 0.05). While other individual enzymes also show trends related to chironomid population density, it is more instructive to consider these enzyme activities on a larger scale. If we consider the mean activity of the enzymes used in carbon acquisition and those used to acquire nitrogen, clear treatment effects are evident (Table 3). For both carbon and nitrogen acquiring enzyme systems activity decreased with increasing chironomid population density. This pattern suggests that the chironomids, through bioturbation, leakage, or physiological processes, produce carbon and nitrogen substrates that are more easily acquired by the microbial community (Gardner et al., 1983; Fukuhara & Yasuda, 1989; Goedkoop, 1993). The activity ratio between carbon and nitrogen acquiring enzymes has been suggested as an indicator of resource allocation by aquatic microbial communities (Christian & Karl, 1995). In this study, the ratio of carbon–nitrogen acquiring enzymes increased with chironomid population density (Table 3). This ratio suggests that although absolute availability of both nutrients increased with chironomid population density, nitrogen was relatively more available than carbon. This may be due to the increase in nitrogen availability through excretion of ammonium by the chironomids
79 (Gardner et al., 1983; Fukuhara & Yasuda, 1989; Chapman, 1989). Substrate utilization Each of the 6 substrate categories derived from the 95 substrates in the GN-plate displayed similar trends. Substrate utilization was greatest in the microcosm with 5 chironomids, and lowest in the 10 chironomid microcosms (Table 4). This pattern suggests enhanced bacterial production with moderate chironomid feeding pressure and repression at higher levels. The lack of a significant difference in bacterial biomass between the 3 treatments implies that this production was the result of the stimulation of either the already active bacterial community, or an increase in ratio of active to dormant bacteria. Traunspurger et al. (1997) noted similar findings in an experiment with bacteria feeding nematodes. They concluded that the grazing affects of nematodes were of more impotents in stimulating bacterial activity than either bioturbation or excretion. This increase in substrate utilization may also be due to an increase in bacterial diversity as a response increases in substrate and chemical heterogeneity at the 5 chironomid population density (Connell, 1978) where the substrate was composed of approximately equal amounts of unused paper toweling, frass and tubes. In the microcosms with 0 and 10 chironomid larvae, the substratum may be more physically and chemical homogeneous decreasing niche breadth and thereby microbial diversity and substrate utilization. This homogeneity was represented by the preponderance of readily identifiable paper toweling in the 0 chironomid treatment, which at least at the visual scale is homogeneous. In the 10 chironomid treatment, the microcosm was dominated by frass, with relatively little identifiable paper toweling most of which was used in tube construction by the chironomids.
Conclusions Using the conventional measure of microbial biomass, this research was unable to detect any effect of chironomid feeding on the microbial community at three levels of feeding pressure by the midge C. tentans. However, a dramatic relationship was observed between microbial function and midge population density. Activities of enzymes used by the microbial community to obtain carbon and nitrogen from the environment were negatively correlated with chironomid population density. This decline in activity suggests that the microbial community was taking advantage of low molecular weight chironomid exudates (organic and inorganic), which would be readily available, and in doing so use less energy to acquire essential nutrients. The carbon to nitrogen enzyme ratio further suggests that nitrogen is more available with increasing chironomid population density relative to carbon. Lastly, as portrayed by cluster analysis, there was a shift microbial community genetic diversity with increasing chironomid population density, with little similarity in the microbial community between microcosms containing larvae and those without. Taken in total, these observations of microbial community structure and function suggest that the relationship between the microbial community and C. tentans is not linear. But, a dynamic interplay where the microbial community, or some product there of, is used as a nutrient source by the chironomid larvae, and the microbes take advantage of chironomid induced physical and/or chemical modifications of the substratum. Future research should focus on further understanding microbial community structure through the use of more definitive molecular techniques, and the precise relationship of microbes as a food source for the chironomids.
Acknowledgements Microbial community heterogeneity Shifts in microbial structure were evident through whole community DNA–DNA hybridizations. The microbial communities at chironomid densities of 5 and 10 chironomids per microcosm were 68 and 100% dissimilar, respectively, to the microbial community in the O chironomid microcosm. While this data suggest a greater number of niches in the presence of the midge larvae, no true measure of diversity can be inferred.
We would lime to thank the Lake Erie Soil and Water Research and Education Center, University of Toledo and Fairmont State College, for its monetary and physical contribution to this study.
References Baird, B. H. & D. Thistle, 1986. Uptake of bacterial extracellular polysaccharides by deposit feeding holothurian (Isostichopus badionotus). Mar. Biol. 92: 183–187.
80 Baker, J. H. & L. A. Bradnam, 1976. The role of bacteria in the nutrition of aquatic detritivores. Eecologia 24: 95–104. Bärlocher, F., and J. H. Murdoch, 1989. Hyporheic biofilms – a potential food source for interstitial animals. Hydrobiologia 184: 61–67. Berg, M. B., 1995. Larval food and feeding behaviour. In Armitage, P. D., P. S. Cranston & L. C. V. Pinder (eds), The Chironomidae: Biology and Ecology of Non-biting Midges. NY, Chapman and Hall: 136–168. Cammen, L. M., 1980. The significance of microbial carbon in the nutrition of the deposit feeding polychaete Nereis succinea. Mar. Biol. 61: 9–20. Carman, K. R. & D. Thistle, 1985. Microbial food partitioning by three species of benthic copepodes. Mar. Biol. 88: 143–148. Chapman, R. F., 1982. The Insects: Structure and Function. Harvard University Press, Cambridge, MA. Christian, J. R. & D. M. Karl, 1995. Bacterial Ectoenzymes in Marine Waters – Activity Ratios and Temperature Responses in 3 Oceanographic Provinces. Limnol. Oceanogr. 40: 1042–1049. Connell, J. H., 1978. Diversity in tropical rain forests and coral reefs. Science 199: 1302–1310. Davies, F. N., 1925. Quantitative studies of the fauna of the sea bottom, No. 2. Results of the investigation in the southern North Sea. Fish Invest. 2, 8. Dean, W. E., 1974. Determination of carbonate and organic matter in clacareous sediments and sedimentary rock by loss on ignition. J. Sed. Petrol. 44: 249–253. Decho, A. W. & G. J. Herndl, 1995. Microbial activities and the transformation of organic matter within mucilaginous material. Sci. Total Environ. 165: 33–42. Findlay, R. H., G. M. King & L. Watling, 1989. Efficacy of phospholipid analysis in determining microbial biomass. Appl. Environ. Microbiol. 55: 2888–2893. Findlay, S., J. L. Meyer & P. J. Smith, 1984. Significance of bacteria biomass in the nutrition of a freshwater isopod (Lirceus sp.). Oecologia 63: 38–42. Fry, J. C., 1982. Interactions between bacteria and benthic invertebrates. In Nedwill, D. B. & C. M. Brown (eds), Sediment Microbiology. Academic Press, NY: 171–201. Fukuhara, H. & K. Yasuda, 1989. Ammonium excretion by some freshwater zoobenthos from a eutrophic lake. Hydrobiologia 173: 1–8. Gardner, W. S., T. F. Nalepa, D. S. Slavenus & G. A. Lairds, 1983. Patterns and rates of nitrogen release by benthic Chironomidae and Oligochaeta. Can. J. Fish. aquat. Sci. 40: 259–266. Garland, J. L. & A. L. Mills, 1991. Classification and Characterization of Heterotrophic Microbial Communities On the Basis of Patterns of Community-Level Sole- Carbon-Source Utilization. Appl. Environ. Microbiol. 57(8): 2351–2359. Garland, J. L., 1997. Analysis and interpretation of community-level physiological profiles in microbial ecology. FEMS Microbiol. Ecol. 24: 289–300. Goedkoop, W. & R. K. Johnson, 1992. Modelling the importance of sediment bacterial carbon for profundal macroinvertebrates along a lake nutrient gradient. Neth. J. Aquat. Ecol. 26: 477–483. Hargrave, B. T., 1970. The effect of a deposit-feeding amphipod on the metabolism of benthic microflora. Limnol. Oceanogr. 15: 21–30. Hargrave, B. T., 1976. The central role of invertebrate feces in sediment decomposition. In Anderson, J. M. & A. Macfayden (eds), The Role of Terrestrial and Aquatic Organisms in Decompostion Process. Blackwell Scientific Publishers, Oxford: 301–321.
Hax, C. L. & S. W. Golladay, 1993. Macroinvertebrate colonization and biofilm development on leaves and wood in a boreal river. Freshwat. Biol. 29: 79–87. Hershey, A. E., R. W. Merritt, M. C. Mille & J. S. McCrea, 1996. Organic matter processing by larval black flies in a temperate woodland stream. Oikos 75: 524–532. Hilsenhoff, W. L., 1966. The biology of Chironomus plumosus (Diptera: Chironomidae) in Lake Winnebago, Wisconsin. Ann. Entomol. Soc. Am. 59: 465–473. Hobbie, J. E. & C. Lee, 1980. Microbial production of extracellular material: Importance in benthic ecology. In Tenore, K. & B. C. Coull (eds), Marine Benthic Dynamics. University of South Carolina Press, University of South Carolina: 341–346. Johnson, R. K., 1984. Distribution of Chironomus plumosus and C. anthracinus with respect to sediment parameters in mesotrophic Lake Erken. Int. Ver. theor. Angew. Limnol. Verh. 22: 750–758 (In Number 1156). Johnson, R. K., B. Bostrom & W. Van De Bund, 1989. Interactions between Chironomus plumosus, L. and the microbial community in surficial sediments of a shallow eutrophic lake. Limnol. Oceanogr. 34: 992–1003. Kemp, P. F., 1987. Potential impact on bacteria of grazing by a macrofaunal deposit-feeder, and the fate of bacteria production. Mar. Ecol. Prog. Ser. 36: 151–161. Lee, S. & J. A. Fuhrman, 1991. Spatial and temporal variation of natural bacterioplankton assemblages studied by total genomic DNA cross-hybridization. Limnol. Oceanogr. 36: 1277–1287. McGarrigle, M. L., 1980. The distribution of chironomid communities and controlling sediment parameters in L. Derravaragh, Ireland. In Murray, D. A. (ed.), Chironomidae: Ecology, Systematics, Cytology, and Physiology. Pergamon Press, Oxford: 275–282. McLachlan, A. J. & C. H. Dickson, 1977. Micro-organisms as a factor in the distribution of Chironomus lugubris Zetterstedt in a bog lake. Arch. Hydrobiol. 80: 133–146. McMurtry, M. J., D. J. Rapport & K. E. Chua, 1983. Substrate selection by Tubificid Oligochaetes. Can. J. Fish. aquat. Sci. 40: 1639–1646. More, M. I., J. B. Herrick, M. C. Silva, W. C. Ghoirse & E. L. Madsen, 1994. Quantitative cell lysis of indigenous microorganisms and rapid extraction of microbial DNA from sediment. Appl. Environ. Microbiol. 60: 1572–1580. Phillips, E. C. & R. V. Kilambi, 1994. Use of course woody debris by diptera in Ozark streams, Arkansas. J. n. am. Benthol. Soc. 13: 151–159. Reynoldson, T. B., 1987. The role of environmental factors in the ecology of tubificid oligochaeted – an experimental study. Hol. Ecol. 10: 241–248. Rodina, A. G., 1949. The role of bacteria in the feeding of tendipedid larvae. Fish. Res. Bd Can. Transl. Series. 1848: 1–6. Sinsabaugh, R. L., M. J. King, H. P. Collins, P. E. Yeager & S. O. Petersen, 1999. Characterizing Soil Microbial Communities. Standard Soil Methods for Long-Term Ecological Research. Robertson, P. D., D. C. Coleman, C. S. Bledsoe & P. Sollins (eds), NY, Oxford University Press: 318–348. Taghon, G. L., 1982. Optimal foraging by deposit-feeding invertebrates: roles of particle size and organic coating. Oecologia 52: 295–304. Toscano, R. J. & A. J. McLachlan, 1980. Chironomids and particles: Microorganisms and chironmid distribution in a peaty upland river. In Murray, D. A. (ed.), Chironomidae-Ecology, Systematics, Cytology, and Physiology. Pergamon Press, Oxford: 171–177.
81 Townsend, B. E., S. G. Lawrence & J. F. Flannagan, 1981. Chironomus tentans Fabricius. Manual for the Culture of Selected Invertebrates. S. G. Lawrence, Can. Special Publ. Fish. aquat. Sci. 54: 109–126. Traunspurger, W., M. Bergtold & W. Goedkoop, 1997. The effects of nematodes on bacterial activity and abundance in a freshwater sediment. Oecologia 112(1): 118–122. Van De Bund, W. J., W. Goedkoop & R. K. Johnson, 1994. Effects of deposit-feeder activity on bacterial production and abundance in profundal lake sediment. J. n. am. Benthol. Soc. 13(4): 532– 539.
Ward, G. M. & K. W. Cummins, 1979. Effects of food quality of growth on a stream detritivore Paratendipes albimanus (Meigen) (Diptera: chironomidae). Ecology 60: 57–64. White, D. C., W. M. Davis, J. S. Nickels, K. J. D. & R. J. Bobbie, 1979. Determination of the sedimentary microbial biomass by extractible lipid phosphate. Oecologia 40: 51–62. Zak, J. C., M. R. Willig, D. L. Moorhead & H. G. Wildman, 1994. Functional Diversity of Microbial Communities – a Quantitative Approach. Soil Biol. Biochem. 26: 1101–1108. ZoBell, C. E. & C. B. Feltham, 1938. Bacteria as food for certain marine invertebrates. J. mar. Res. 1: 312–327.
